--- Comment #5 from thomas-carsten dot franke at brunel dot de 2010-06-18
06:40 ---
Created an attachment (id=20937)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=20937&action=view)
extract of .s file (-save-temps)
shows the lpm r31,Z, only
--
http://gcc.gnu.org/bugzilla/sho
--- Comment #4 from thomas-carsten dot franke at brunel dot de 2010-06-18
06:38 ---
Created an attachment (id=20936)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=20936&action=view)
extract of *.lss file
shows the Z+ in lpm
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=4456
--- Comment #3 from eric dot weddington at atmel dot com 2010-06-18 03:20
---
The code you have in the bug report shows:
5808: f4 91 lpm r31, Z+
Take a long hard look at the hex instruction. Reverse the bytes: 91 F4
The LPM instruction in the instruction set docume
--- Comment #2 from eric dot weddington at atmel dot com 2010-06-18 03:05
---
Please attach a preprocessed test case, and the generated assembly (the .S file
when you use -save-temps).
--
eric dot weddington at atmel dot com changed:
What|Removed
--- Comment #9 from pinskia at gcc dot gnu dot org 2010-06-18 00:49 ---
(In reply to comment #8)
> Can we use subreg instead of vec_select?
Kinda, you need to do triple subregs, first to an integer mode and then to a
smaller integer mode and then to the other vector mode. subreg on vec
--- Comment #8 from hjl dot tools at gmail dot com 2010-06-18 00:46 ---
Can we use subreg instead of vec_select?
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=44551
--- Comment #19 from danglin at gcc dot gnu dot org 2010-06-18 00:32
---
Fixed.
--
danglin at gcc dot gnu dot org changed:
What|Removed |Added
Status|UNCONFI
--- Comment #18 from danglin at gcc dot gnu dot org 2010-06-18 00:32
---
Subject: Bug 43740
Author: danglin
Date: Fri Jun 18 00:31:53 2010
New Revision: 160958
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=160958
Log:
PR target/43740
config/pa/pa.c (emit_move_s
--- Comment #17 from danglin at gcc dot gnu dot org 2010-06-18 00:29
---
Subject: Bug 43740
Author: danglin
Date: Fri Jun 18 00:29:30 2010
New Revision: 160957
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=160957
Log:
PR target/43740
config/pa/pa.c (emit_move_s
--- Comment #16 from danglin at gcc dot gnu dot org 2010-06-18 00:26
---
Subject: Bug 43740
Author: danglin
Date: Fri Jun 18 00:26:33 2010
New Revision: 160956
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=160956
Log:
PR target/43740
config/pa/pa.c (emit_move_s
--- Comment #15 from danglin at gcc dot gnu dot org 2010-06-18 00:24
---
Subject: Bug 43740
Author: danglin
Date: Fri Jun 18 00:23:51 2010
New Revision: 160955
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=160955
Log:
PR target/43740
config/pa/pa.c (emit_move_s
$ cat vararg.c
#include
#include
#include
int fails = 0;
struct S116 { float a[3]; } ;
struct S116 a116[5];
void check116va (int z, ...)
{ struct S116 arg, *p;
va_list ap;
int j=0,k=0;
int i;
__builtin_va_start(ap,z);
for (i = 2; i < 4; ++i) {
p = NULL;
j++;
k+=2;
swi
--- Comment #10 from spop at gcc dot gnu dot org 2010-06-17 23:33 ---
Fixed.
--
spop at gcc dot gnu dot org changed:
What|Removed |Added
Status|UNCONFIRMED
--- Comment #8 from burnus at gcc dot gnu dot org 2010-06-17 23:13 ---
Created an attachment (id=20935)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=20935&action=view)
Draft patch, which keeps diagnostic for derived types
Draft patch, which fixes this regression and also adds a c
--- Comment #3 from janus at gcc dot gnu dot org 2010-06-17 22:59 ---
(In reply to comment #1)
> The modified code compile with 4.5.0, hence
> a regression I can trace between revisions 158215 (working) and 158921 (ICE).
The regression is most probably due to the merging of the fortra
--- Comment #1 from janus at gcc dot gnu dot org 2010-06-17 22:46 ---
This is not expected to work right now, as polymorphic arrays are not really
implemented yet.
--
janus at gcc dot gnu dot org changed:
What|Removed |Added
---
The atoi function has undefined behavior if its argument is outside
the range of int. Thus, GCC should not use it for any user input that
might be outside that range.
For example, -O4294967296 should act like -O3, but on 64-bit systems with
glibc it acts like -O0 (glibc's atoi converts the return
--- Comment #1 from janus at gcc dot gnu dot org 2010-06-17 22:38 ---
*** This bug has been marked as a duplicate of 43969 ***
--
janus at gcc dot gnu dot org changed:
What|Removed |Added
--
--- Comment #8 from janus at gcc dot gnu dot org 2010-06-17 22:38 ---
*** Bug 44570 has been marked as a duplicate of this bug. ***
--
janus at gcc dot gnu dot org changed:
What|Removed |Added
---
--- Comment #2 from janus at gcc dot gnu dot org 2010-06-17 22:34 ---
Reduced test case:
module ice6
type :: t
contains
procedure :: get_array
generic :: get_something => get_array
end type
contains
function get_array(this)
class(t) :: this
real,dimension(
--
janus at gcc dot gnu dot org changed:
What|Removed |Added
Status|UNCONFIRMED |NEW
Ever Confirmed|0 |1
Keywor
--- Comment #4 from janus at gcc dot gnu dot org 2010-06-17 22:18 ---
Fixed with r160948. Closing.
Thanks for the report!
--
janus at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #3 from janus at gcc dot gnu dot org 2010-06-17 22:15 ---
Subject: Bug 44558
Author: janus
Date: Thu Jun 17 22:15:30 2010
New Revision: 160948
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=160948
Log:
2010-06-17 Janus Weil
PR fortran/44558
* reso
--- Comment #7 from hjl dot tools at gmail dot com 2010-06-17 22:01 ---
Created an attachment (id=20934)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=20934&action=view)
A patch to split cast
Here is a patch to split cast. But it doesn't remove
redundant vinsertf128/vextractf128.
--- Comment #26 from bernds at gcc dot gnu dot org 2010-06-17 21:54 ---
Fixed.
--
bernds at gcc dot gnu dot org changed:
What|Removed |Added
Status|NEW
--- Comment #25 from bernds at gcc dot gnu dot org 2010-06-17 21:52 ---
Subject: Bug 39871
Author: bernds
Date: Thu Jun 17 21:51:55 2010
New Revision: 160947
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=160947
Log:
PR rtl-optimization/39871
* reload1.c (init_el
--- Comment #2 from redi at gcc dot gnu dot org 2010-06-17 19:57 ---
The program has undefined behaviour because it declares a name in namespace
std, which is reserved for the implementation. You can workaround this as
Andrew suggests, but the behaviour is still technically undefined.
--- Comment #1 from lists at eitanadler dot com 2010-06-17 19:35 ---
Created an attachment (id=20933)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=20933&action=view)
preprocessed source of file that fails
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=44573
This is reproducible on multiple computers (both with the same uname -rms)
which rules out hardware errors.
[ei...@voodoo ~]$ uname -rms
FreeBSD 8.1-PRERELEASE i386
[ei...@voodoo ~]$ gcc -v
Using built-in specs.
Target: i386-undermydesk-freebsd
Configured with: FreeBSD/i386 system compiler
Thread
--- Comment #2 from jakub at gcc dot gnu dot org 2010-06-17 19:18 ---
Hopefully fixed.
--
jakub at gcc dot gnu dot org changed:
What|Removed |Added
Status|UNC
--- Comment #1 from jakub at gcc dot gnu dot org 2010-06-17 19:16 ---
Subject: Bug 44572
Author: jakub
Date: Thu Jun 17 19:15:51 2010
New Revision: 160944
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=160944
Log:
PR debug/44572
* dwarf2out.c (dwarf2out_debug_hoo
--- Comment #7 from kargl at gcc dot gnu dot org 2010-06-17 19:08 ---
The following patch restores the 4.4.0 behavior for
STAT of not checking derived type components. This
of course now allows invalid code to compile such as
this modified version of OP's test subroutine
subroutine
--- Comment #6 from ceggers at gmx dot de 2010-06-17 18:56 ---
(In reply to comment #5)
> Do we need to act as if
> -fno-ira-share-spill-slots
> is set in cfun->calls_setjmp functions?
At least in my case "-Os -fno-ira-share-spill-slots" seems to solve the
problem. This applies also to
/test/gnu/gcc/objdir/./prev-gcc/xgcc -B/test/gnu/gcc/objdir/./prev-gcc/
-B/opt/g
nu/gcc/gcc-4.6.0/hppa2.0w-hp-hpux11.11/bin/
-B/opt/gnu/gcc/gcc-4.6.0/hppa2.0w-hp
-hpux11.11/bin/ -B/opt/gnu/gcc/gcc-4.6.0/hppa2.0w-hp-hpux11.11/lib/ -isystem
/op
t/gnu/gcc/gcc-4.6.0/hppa2.0w-hp-hpux11.11/include -isyst
--- Comment #1 from pinskia at gcc dot gnu dot org 2010-06-17 18:47 ---
std::strcmp is a builtin if you don't want to use the builtin function, use
either -fno-builtins or -fno-builtin-strcmp.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=44571
The code below complains that strcmp is not a member of std. It does not
complain about the function strcmp_1.
The same code compiled correction on gcc 4.1.2.
This was compiled with gcc -c.
extern int strcmp( const char* s, const char* s2 );
extern int strcmp_1( const char* s, const char* s2
--- Comment #1 from wvangulik at xs4all dot nl 2010-06-17 18:38 ---
The linked mail shows that 4.2.2 in that test case did not suffer this problem
(it uses lpm r31, Z) which is ok.
http://lists.nongnu.org/archive/html/avr-gcc-list/2008-01/msg00026.html
Maybe it is a xmega specific thing
--- Comment #13 from eblot dot ml at gmail dot com 2010-06-17 16:41 ---
(In reply to comment #10)
> When I use -fshort-enums my dumps specifically state unsigned char.
> But I assume that char is actually unsigned here by default.
AAPCS specifies the following (7.1.1)
C/C++ type: M
--- Comment #12 from jamborm at gcc dot gnu dot org 2010-06-17 16:37
---
(In reply to comment #11)
> (In reply to comment #10)
> > > With -fno-tree-switch-conversion switch option switch, the bug does not
> > > appear.
> > Really? See below...
>
> Below is the dump WITHOUT the `-fno-
--- Comment #11 from eblot dot ml at gmail dot com 2010-06-17 16:34 ---
(In reply to comment #10)
> > With -fno-tree-switch-conversion switch option switch, the bug does not
> > appear.
> Really? See below...
Below is the dump WITHOUT the `-fno-tree-switch-conversion`
I did not unders
This one might be related to bug 44568 since it is also related to polymorphic
derived type arrays.
module ice8
type::a_type
end type a_type
contains
subroutine do_something()
class(a_type),dimension(:),allocatable::values
if(allocated(values))deallocate(values)
end subroutine
--- Comment #6 from kargl at gcc dot gnu dot org 2010-06-17 15:26 ---
(In reply to comment #5)
> Remove [4.5/4.6 Regression] from the summary as this
> has never worked, so it can't be a regression.
>
Note, the OP's code appears to work in 4.4.0 because
prior to the commit noted in co
--- Comment #5 from kargl at gcc dot gnu dot org 2010-06-17 15:22 ---
Remove [4.5/4.6 Regression] from the summary as this
has never worked, so it can't be a regression.
--
kargl at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #6 from amylaar at gcc dot gnu dot org 2010-06-17 15:20 ---
Subject: Bug 44512
Author: amylaar
Date: Thu Jun 17 15:20:20 2010
New Revision: 160922
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=160922
Log:
PR bootstrap/44512
* c-cppbuiltin.c (builtin_
--- Comment #10 from jamborm at gcc dot gnu dot org 2010-06-17 15:00
---
(In reply to comment #7)
> (In reply to comment #5)
> > Could you please first try to reproduce the bug with the
> > -fno-tree-switch-conversion swithch?
>
> Using GCC 4.5.0 (tarball)
>
> With -fno-tree-switch-c
--- Comment #2 from joseph at codesourcery dot com 2010-06-17 14:28 ---
Subject: Re: New: configuration with multiple targets
/ backends is not supported.
On Thu, 17 Jun 2010, amylaar at gcc dot gnu dot org wrote:
> To a first order of approximation, we can say trees and their passe
--- Comment #5 from amylaar at gcc dot gnu dot org 2010-06-17 14:12 ---
A patch set is here:
http://gcc.gnu.org/ml/gcc-patches/2010-06/msg01765.html
http://gcc.gnu.org/ml/gcc-patches/2010-06/msg01766.html
--
amylaar at gcc dot gnu dot org changed:
What|Removed
I am working on a vector shifting patch,
http://gcc.gnu.org/ml/gcc-patches/2010-06/msg01511.html and I noticed the
following effect. When I compile the code with "-O -g", then the compiler
crashes with an error:
"internal compiler error: in simplify_subreg, at simplify-rtx.c:5128".
If I use -On o
--- Comment #9 from jamborm at gcc dot gnu dot org 2010-06-17 14:02 ---
(In reply to comment #8)
> Subject: Re: switch/case optimization produces an invalid lookup table index
>
> I bet it could be reproduced on any target with -fshort-enums.
>
Unfortunately no, this switch does not
--- Comment #3 from mikpe at it dot uu dot se 2010-06-17 13:51 ---
The ICE was introduced in 4.5 by r146904, an ira tweak for a missed
optimization (PR39914) with no ARM specific bits. That change was then applied
to 4.4 in r147081, and indeed that caused 4.4 branch to ICE too.
I'll bi
--- Comment #3 from zsojka at seznam dot cz 2010-06-17 13:51 ---
It started crashing between r160527 and r160687
--
zsojka at seznam dot cz changed:
What|Removed |Added
module ice6
type::a_type
contains
procedure::do_something
end type a_type
contains
subroutine do_something(this)
class(a_type),intent(in)::this
end subroutine do_something
subroutine do_something_else()
class(a_type),dimension(:),allocatable::values
call values(1
Broken on 6.x gch45 core dumps on 6.x with no immediately obvious fix
--
Summary: FreeBSD 6.x compile error
Product: gcc
Version: unknown
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: libgcj
Assigned
--- Comment #5 from jamborm at gcc dot gnu dot org 2010-06-17 13:14 ---
A request for approval to backport the two patches sent to the mailing list:
http://gcc.gnu.org/ml/gcc-patches/2010-06/msg01663.html
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=44133
--- Comment #14 from dominiq at lps dot ens dot fr 2010-06-17 12:30 ---
>From comments #11 to #13, given links and
http://gcc.gnu.org/ml/gcc-patches/2010-06/msg01744.html closing as fixed.
--
dominiq at lps dot ens dot fr changed:
What|Removed |Add
--- Comment #4 from jakub at gcc dot gnu dot org 2010-06-17 12:28 ---
Created an attachment (id=20932)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=20932&action=view)
callgrind.out.bz2
Seems it is mainly df and RA that eats the time for -O0 -g0.
--
http://gcc.gnu.org/bugzill
--- Comment #1 from amylaar at gcc dot gnu dot org 2010-06-17 11:40 ---
(In reply to comment #0)
> multi-target configurations, thus PR44362 doesn't actually block.
Sorry, I meant PR44431. But then, the remaining part of PR44362 also
doesn't block PR44566 (and not PR44433 either).
--
As ramping up clock speeds becomes harder and we see more parallel cores
being used to extend Moore's Law, there is also a growing number of
heterogenous
systems that have fully programmable cores with different architectures, e.g.
Cell, ARCompact / MXP, CPU / GPU combinations.
It is desirable tha
--- Comment #1 from dominiq at lps dot ens dot fr 2010-06-17 11:07 ---
Confirmed on thrunk after having added
integer :: dim = 2
in 'type :: lin_approx_cont_type'. The modified code compile with 4.5.0, hence
a regression I can trace between revisions 158215 (working) and 158921 (
--- Comment #3 from rguenth at gcc dot gnu dot org 2010-06-17 10:44 ---
Growth in time and memory is linear in the number of functions for me (GCC
4.5).
rguent...@murzim:/tmp> ~/bin/maxmem2.sh /usr/bin/time gcc-4.5 -S gcc_16kgen1.c
-o /dev/null
14.60user 0.61system 0:15.42elapsed 98%CPU
--- Comment #2 from rguenth at gcc dot gnu dot org 2010-06-17 10:36 ---
The issue is not so much the number of functions but the number of calls
in main (and thus its size).
With GCC 4.3.4 I see
parser: 6.87 (12%) usr 0.79 (14%) sys 8.14 (13%) wall
266316 kB (18
module ice6
type :: lin_approx_cont_type
real,dimension(:),allocatable::values
contains
procedure::get_array
procedure::get_element
generic::get_something=>get_element,get_array
end type lin_approx_cont_type
contains
pure function get_element(this,index)
class(
--- Comment #4 from burnus at gcc dot gnu dot org 2010-06-17 10:25 ---
For completeness, the check was introduced for PR 38389 by the commit
http://gcc.gnu.org/ml/gcc-cvs/2009-03/msg00838.html
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=44556
--- Comment #8 from pinskia at gmail dot com 2010-06-17 09:37 ---
Subject: Re: switch/case optimization produces an invalid lookup table index
I bet it could be reproduced on any target with -fshort-enums.
On Jun 17, 2010, at 2:20 AM, "eblot dot ml at gmail dot com"
wrote:
>
>
> ---
I bet it could be reproduced on any target with -fshort-enums.
On Jun 17, 2010, at 2:20 AM, "eblot dot ml at gmail dot com" > wrote:
--- Comment #6 from eblot dot ml at gmail dot com 2010-06-17
09:20 ---
(In reply to comment #5)
Unfortunately I don't see this happening on the x8
--- Comment #7 from eblot dot ml at gmail dot com 2010-06-17 09:34 ---
(In reply to comment #5)
> Could you please first try to reproduce the bug with the
> -fno-tree-switch-conversion swithch?
Using GCC 4.5.0 (tarball)
With -fno-tree-switch-conversion switch option switch, the bug do
--- Comment #6 from eblot dot ml at gmail dot com 2010-06-17 09:20 ---
(In reply to comment #5)
> Unfortunately I don't see this happening on the x86_64-linux ->
> arm-linux-gnueabi cross compiler I built for myself. The generated
> assembly has all four elements in the initialization o
--- Comment #5 from manu at gcc dot gnu dot org 2010-06-17 08:37 ---
(In reply to comment #4)
> It seems that optimizing is what's causing the problem: the example compiles
> fine with -O0, but not -On>=1. It also compiles fine when the case values are
> consecutive, which seems telling.
--- Comment #4 from anthony dot penniston at hotmail dot com 2010-06-17
08:16 ---
It seems that optimizing is what's causing the problem: the example compiles
fine with -O0, but not -On>=1. It also compiles fine when the case values are
consecutive, which seems telling. My first guess w
When optimization is enabled during build, the following switch statement is
compiled into assembler using a command combination (lpm r31, Z+) which
potentially results in an undefined result (see LPM description in AVR 8-bit
instruction set manual):
// successfully decoded
switch
--- Comment #5 from manu at gcc dot gnu dot org 2010-06-17 07:29 ---
FIXED in GCC 4.6. Thanks for the report.
--
manu at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #4 from manu at gcc dot gnu dot org 2010-06-17 07:28 ---
Subject: Bug 44486
Author: manu
Date: Thu Jun 17 07:28:21 2010
New Revision: 160877
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=160877
Log:
2010-06-17 Manuel López-Ibáñez
PR c++/44486
cp/
--- Comment #3 from manu at gcc dot gnu dot org 2010-06-17 07:18 ---
You are right. The issue occurs in VRP but not because of the disjoint ranges.
Pass vrp1 is able to optimize your first example (nested if) but not the second
(nested switch). I think this is a missed optimization. Not
73 matches
Mail list logo
